Sampson led the creation of the Department of Statistics

Allan R. Sampson, who was instrumental in establishing the Department of Statistics and was its founding chair (1997-2000), died Jan. 30, 2021.

“He had a reputation for being the tough but fair teacher,” noted current Chair Satish Iyengar, who arrived as a faculty member in the former Department of Mathematics and Statistics in 1982. “As a mentor, he was just fabulous. Alan has had a history of very high-quality mentoring. He has put his students on a very strong track … (with) a strong history of advising Ph.D. students who have done exceptionally well after graduating.”

Sampson’s efforts were recognized with the provost’s award for mentoring at Pitt.

“He introduced me to things that afforded me opportunities,” Iyengar recalled. “You could tell that he was interested in promoting other faculty careers.”

Most recently, Sampson held workshops for graduate students on the do’s and don’ts of interviewing, helping in particular the department’s many foreign-born students with their acculturation to the American system of recruitment and hiring, Iyengar said. Sampson took a very hands-on approach to this work, connecting students with industry representatives and even helping them to write their resumes.

But as Sampson’s former departmental colleague Leon Gleser noted at the remembrance gathering on Feb. 19, creating a separate statistics department at Pitt was a struggle that spanned nearly a decade.

“His plan convinced the administration that there would be only a minor additional cost to their budget from establishing a department,” Gleser said of Sampson. “His leadership and diplomacy were crucial in establishing momentum for the department…  The only problem with Allan’s leadership was that it was so good that when the time came to elect a new chair, no one volunteered.”

Born Aug. 25, 1945, Sampson received his bachelor’s degree in mathematics from UCLA, where he was a member of Phi Beta Kappa, and his master’s degree (1968) and doctorate (1970) in statistics from Stanford.

Sampson began his academic career in 1970 as an assistant professor of statistics at Florida State University but immediately began expanding his experience as a visiting lecturer in statistics and operations research at Tel Aviv University in Israel.

From 1975 through 1978, he turned to industry, working as manager of the Department of Biostatistics, Pharmaceutical Products Division, of Abbott Laboratories, until he joined Pitt as an associate professor. During his years at the University, he was also a visiting professor or scholar at Carnegie Mellon University, University of California–San Diego, Stanford and Tel Aviv.

Sampson was elected as a fellow of all the significant organizations in his profession, Iyengar said, including the American Statistical Association and the Institute of Mathematical Statistics. He also received the Dissertation Summary Award from the Drug Information Journal in 1993, a $20,000 prize given to him and the student he had mentored.

His initial research focus on multivariate analysis took a more practical turn as his career progressed, Iyengar said, thanks to his experience in industry. Sampson’s career was filled with years of work advising committees of the Food and Drug Administration on the statistics as applied to multiple diseases and disabilities, and finally advised them on hiring statisticians for such work. Perhaps inspired by his use of a wheelchair following an early bout of polio, Sampson also advised federal and local groups on disability issues.

He served as editorial board member or editor for half a dozen journals and wrote numerous research papers, following the receipt of dozens of grants. His work often focused on practical applications of statistics, Iyengar pointed out, including to the study of gun violence. “Technically he was really strong, but also he understood the potential role of statistics is important in public policy questions,” he said.

Sampson’s legacy continues today, Iyengar added. In the 1990s, Sampson started a statistical consulting service in the department as a graduate student class, which he, Iyengar and others taught.

“We would put posters around campus — if you need statistical consulting, we are a resource,” Iyengar recalled. Researchers from the law school, linguistics, health sciences and elsewhere came to the class, were assigned a graduate student as a consultant and presented their needs to the students.

“It was an excellent training device,” Iyengar said. “We served, and we still do, the research community broadly. It is free to the Pitt community.”

Young researchers who may have grant funding for a research study, but still can’t afford to hire a statistician, can take advantage of the program today.

“This is one of Alan’s great legacies in the department,” Iyengar said.

— Marty Levine